#!/usr/bin/env python
# coding: utf-8

# # Loading DICOM images into PyTorch
# 
# > Recently I needed to load DICOM files for modeling and I've learned about handly library - imageio. It can handle a lot of different file types effortlessly, one of them DICOM volumetrtic datasets. 
# 
# - toc: false
# - hide: false
# - badges: true
# - comments: true
# - categories: [ML, medical]

# In[1]:


# install imageio if you haven't already
get_ipython().system('pip install imageio')


# In[2]:


import imageio
import torch


# As an example, I've downloaded some dicom files from [this site](https://www.visus.com/en/downloads/jivex-dicom-viewer.html)

# In[3]:


get_ipython().system('curl "https://www.visus.com/fileadmin/content/pictures/Downloads/JiveX_DICOME_Viewer/case1.zip" > "case1.zip"')

get_ipython().system('unzip -q case1.zip')


# and simply pass the folder to imageio like this:

# In[4]:


np_arr = imageio.volread('case1')


# ### turn into torch tensor
# As I prefert to work with PyTorch tensors...

# In[5]:


dicom_torch = torch.from_numpy(np_arr)


# In[6]:


dicom_torch.shape


# ... and this is how it looks like

# In[7]:


import matplotlib.pyplot as plt
get_ipython().run_line_magic('matplotlib', 'inline')


# In[8]:


plt.imshow(dicom_torch[10])